144 CARNEGIE INSTITUTION OF WASHINGTON. 



PUBLICATIONS. 

 Brief reviews of the papers published by members of the Laboratory 

 staff during the current year follow. 



(1) The position of the vibration plane of the polarizer in the petrographic microscope. 



F. E. Wright. J. Wash. Acad. Sci., 5, 641-644 (1915). 



In petrographic microscopes the polarizer is mounted in one of two positions, 

 so that its plane of light transmission is parallel either to the vertical cross-hair 

 or to the horizontal cross-hair of the eyepiece. In case the incident light be 

 non-polarized both positions of the polarizer are equally good, as they transmit 

 equal amounts of light. If, however, the incident light be partially polarized, 

 that position of the polarizer which transmits the most light is obviously the 

 better. Partial polarization arises from two factors: (1) Reflection by the 

 surface of the substage mirror. This is not serious, because under ordinary 

 conditions less than 10 per cent of the light thus reflected is plane-polarized. 



(2) Sky polarization. This varies with the line of sight and is at a maximum 

 in the plane polar to the sun. On a clear day from 40 to 80 per cent of the 

 light is plane-polarized in this plane, which is also the plane of vibration of the 

 polarized rays. As this plane moves with the sun, the plane of vibration of 

 rays incident on a microscope facing north is nearly vertical in the early morn- 

 ing and later afternoon hours, while at noon it is horizontal. The optimum 

 position of the polarizer varies, therefore, with the time of day at which observa- 

 tions are made. In case the polarizer is fixed in position, its plane of vibration 

 should be parallel to the vertical cross-hair of the eyepiece. 



(2) The later stages of the evolution of the igneous rocks. N. L. Bowen. J. Geol., 23, 

 Suppl., 1-91 (1915). 



With the continuance of experimental studies of silicate melts, systems have 

 been investigated which approach in their complexity some of the simpler 

 types of igneous rocks. As a result, new light is thrown on some of the 

 problems of igneous-rock genesis, and in this paper an attempt has been made 

 to discuss these problems with the emphasis placed on the significance of the 

 experimental studies. 



Of all the processes that have been suggested as important factors in the 

 differentiation of rocks, crystallization alone seems to be thoroughly compe- 

 tent to produce large results. The formation of various rock types through 

 the agency of crystallization depends on the fact that the various minerals in 

 a fused rnixture separate in a definite order, and that during the period of 

 crystallization an accumulation of crystals of a certain kind may take place 

 in one part of the mass with corresponding impoverishment of this material 

 in other parts. The accumulation of crystals is believed to take place largely 

 through the action of gravity. Experimental evidence shows definitely that 

 the sinking of crystals in silicate melts can take place and a variety of field 

 observations indicate its importance in nature. 



The crystallization of mixtures containing a number of important rock- 

 forming silicates has been very thoroughly studied. These silicates include, 

 among others, the plagioclases, pyroxenes, olivines, spinel, and silica in its 

 various forms, and the order of their appearance for various mixtures has been 

 definitely ascertained. The order shows plainly that, under the action of 

 gravity, there would be a definite tendency towards the grouping of the 

 olivines, pyroxenes, spinel, and the basic plagioclases in the differentiates of 

 early consolidation, with a resultant enrichment of the later differentiate in 

 alkaline feldspars and free silica. This is the kind of differentiates we find in 

 nature in the gabbro-diorite-granite sequence. 



